Purpose: Little is known about the dispersion of a defined contrast bolus during its passage through the heart and pulmonary vasculature. The purpose of this study was to analyze factors influencing a defined contrast bolus for ce-MRA of thoracic vessels.
Materials and methods: For analysis of bolus geometry, an ECG-gated saturation-recovery Turbo-Flash sequence with a TI of 20 msec was used. It was acquired axially at the level of the pulmonary trunc, so that with one data acquisition a curve analysis was possible in the ascending and descending aorta, and in the pulmonary trunc. Twenty-nine patients received 3 ml of Gd-DTPA diluted with saline to a total of 20 ml. Contrast injection was done using a MR compatible power injector with injection rates varying between 1, 2 and 4 ml/sec. Each injection was followed by a saline flush of 20 ml with the same injection rate and mode. Cardiac function was assessed by cine imaging, and phase contrast measurements. After normalization to baseline signal intensity (SI), bolus curves were fitted using a gamma-variate fit and peak signal intensity (peak SI), time-to-peak (TP), upslope, mean transit time (MTT) and dispersion of the contrast bolus were calculated. Furthermore, T (1) and [Gd] in the experimental setting were calculated as follows: T (1) = T (1 o)/ ln [SI/SI (0)], and [Gd] (exp) = [1/T (1) - 1/T (1 o)]/ R (1.) They were then extrapolated [Gd] to clinical conditions by [Gd] (clin) = [Gd] (exp) . 10/1.5, and minimal blood T (1) by T (1)(clin) = 1 / [1/T (1 o) + R (1) [Gd] (clin)].
Results: With increasing injection rate, there was a significant decrease (p < 0.001) of MTT in all target vessels. However, this decrease was not linear: a 4-fold increase in injection rate lead to a 2-fold decrease in MTT e. g. in the ascending aorta. MTT was significantly shorter in the pulmonary trunc compared with that in the ascending and descending aorta (p < 0.001), regardless of injection rate (p < 0.001). Vice versa, dispersion of the contrast bolus was significantly lower in the pulmonary trunc, and increased with higher injection rates. There was no clinically relevant difference in minimal blood T (1) between the different target vessels, for clinical conditions extrapolated values ranged between 20 und 79 msec. Heart function parameters only had a minor influence of bolus curve parameters.
Conclusion: Analysis of bolus geometry enables determination of transit times of a defined contrast bolus through a defined target vessel in the thoracic cavity. Bolus geometry is mainly determined by injection parameters, cardiac function is of minor importance. Dispersion of contrast bolus and MTT increase from the pulmonary trunc to the ascending aorta. The knowledge of these facts may help optimizing of injection parameters and the total amount of contrast agent for contrast-enhanced MRA of thoracic vessels.